Visual Function, Ocular Morphology and Growth – Children Born Moderate-to-
Late Preterm
LINA RAFFA, MD, FEBO
Department of Clinical Neuroscience Institute of Neuroscience and Physiology
The Sahlgrenska Academy at the University of Gothenburg, Gothenburg, Sweden
Gothenburg, 2016
Visual Function, Ocular Morphology and Growth – Children Born Moderate-to-Late Preterm
© LINA RAFFA, MD, FEBO 2016 lina_raffa@yahoo.com
ISBN 978-91-628-9913-4 (PRINT) ISBN 978-91-628-9914-1 (PDF) http://hdl.handle.net/2077/44854
Published articles have been reprinted with permission of the copyright holder.
Printed by INEKO AB, Gothenburg, Sweden, 2016
“Seek knowledge from the cradle to the grave”
Prophet Mohammed (PBUH)
Children Born Moderate-to-Late Preterm LINA RAFFA, MD, FEBO
Department of Clinical Neuroscience, Institute of Neuroscience and Physiology
The Sahlgrenska Academy at the University of Gothenburg, Gothenburg, Sweden
ABSTRACT
Introduction: In the past, researchers have closely studied both systemic and ophthalmological complications associated with extreme preterm birth. Moderate-to-late preterm (MLP) infants have become the fastest-growing subgroup of preterm infants in the last decade, accounting for 84% of all preterm births. Evidence is currently emerging that even near-term birth predisposes those children to a higher risk of mortality and morbidity than term infants. Effects of extreme prematurity on ocular development are known to include retinopathy of prematurity (ROP), refractive errors, strabismus, low visual performance, decreased contrast sensitivity, visual field defects, colour vision deficits and abnormal cognitive development. To date, very few studies have focused on the ophthalmological aspects of this particular subset of MLP children. The aim of the project was to investigate the development of ocular morphology and visual function in children born MLP, relating them to auxological data and comparing them with their full-term counterparts.
Methods: In a prospective population-based study conducted in 2002-2004, 247 potentially eligible children (110 girls and 137 boys) born MLP (gestational age (GA) 32-36 weeks) participated in the neonatal study. None of the participating children had a previous history of ROP. At 5.5, 8 and 12 years of age, 78, 50 and 22 children respectively who were still included in the study took part in sub-studies that focused on orthoptic evaluation, ocular morphology, visual function and electrophysiology in relation to auxological data in both MLP and sex- and age-matched controls.
Results: Based on our findings, being born MLP is associated with increased ocular morbidity and may require greater ophthalmic surveillance than full-term counterparts. Auxological data at birth, especially birth weight, seems to be an important risk indicator when establishing an ophthalmological diagnosis in preschool MLP children, and visual acuity outcome was positively correlated to GA. Good catch-up growth favoured proper development of ocular growth and morphology. Our results show that macular morphology, visual evoked potential (VEP) and full-field electroretinography (ff-ERG) responses are also affected in the MLP group at 12 years of age.
Conclusion: It has been confirmed in our study that preterm birth, even just in the moderate to late phase, represents a continuum of risks associated with visual system morbidities. These findings have potentially important implications for the follow-up of premature children and therefore require confirmation in large population-based studies that encompass these MLP premature children.
Keywords: Auxological data, Electroretinography, IGF-I, Moderate-to-late preterm, Ocular growth, Optical coherence tomography, Retinal nerve fibre layer, Visual evoked potential, Visual function.
ISBN: 978-91-628-9913-4 (PRINT) 978-91-628-9914-1 (PDF)
SAMMANFATTNING PÅ SVENSKA
Introduktion: Forskare har studerat både systemiska och oftalmologiska komplikationer i samband med extremt tidig födsel. Måttligt prematurfödda spädbarn (Moderate-to-late preterm -MLP) har blivit den snabbast växande undergrupp av för tidigt födda barn under det senaste decenniet, och motsvarar 84% av alla prematura födslar. Nya forskningsresultat tyder på att även barn födda måttligt för tidigt är predisponerade för en ökad risk för dödlighet och sjuklighet jämfört med fullgångna barn. Effekter av mycket för tidig födelse avseende utveckling av ögon och synfunktion är kända. Det inkluderar prematuritetsretinopati (ROP), brytningsfel, skelning, minskad kontrastkänslighet, försämrad synskärpa, synfältsdefekter, påverkat färgseende och onormal kognitiv utveckling. Hittills har mycket få studier fokuserat på oftalmologiska aspekter av denna speciella undergrupp av barn födda måttligt för tidigt.
Syftet med projektet var att undersöka och följa utvecklingen av ögonmorfologi och synfunktion hos barn födda MLP; relatera dem till auxologiska data vid födsel respektive vid olika åldrar och jämföra dem med fullgångna kontroller.
Metoder: I en prospektiv populationsbaserad studie i Göteborgsområdet, 247 potentiellt möjliga barn (110 flickor och 137 pojkar) födda MLP (gestationsålder 32-36 veckor) under åren 2002-2004 tackade ja till den neonatala delen av studien. Ingen av de deltagande barnen hade tidigare anamnes av ROP. Barnen följdes upp vid 5.5, 8 och 12 års ålder, och då deltog 78, 50 och 22 barn respektive med fokus på ortoptisk utvärdering, ögonmorfologi, synfunktion och elektrofysiologi i förhållande till auxologiska data i båda MLP och köns- och åldersmatchade kontroller.
Resultat: Att födas måttligt för tidigt visade sig vara förknippat med ökad ögon sjuklighet och kräver således en noggrannare kontroll av ögon och synfunktion jämfört med fullgångna barn.
Auxologiska uppgifter vid födseln, särskilt födelsevikt, tycks vara en viktig riskindikator för en ögondiagnos hos måttligt för tidigt födda förskolebarn. Resultaten av synskärpa var positivt korrelerad till födelsevecka. God ”catch-up” avseende tillväxten tycks gynna utveckling av ögontillväxt och morfologi. Våra resultat visar också att morfologi i området för gula fläcken (makula) och funktionsutfall mätta med visual evoked potential (VEP) och full-fälts elektroretinogram (ff-ERG) var påverkade hos de måttligt för tidigt födda barnen vid 12 års ålder.
Slutsats: Våra resultat visar på att prematuritet, även i måttlig till sen fas, representerar ett kontinuum av risker i samband med sjuklighet inte enbart systemisk, utan även i öga och synsystem. Dessa fynd har potentiellt viktiga konsekvenser för uppföljning av måttligt för tidigt födda barn och kräver därför ytterligare studier för att få en bättre förståelse för mekanismerna bakom dessa processer.
LIST OF PAPERS
This thesis is based on the following studies, referred to in the text by their Roman numerals.
I. Lina H. Raffa, Ann Hellström, Eva Aring, Susann Andersson, Marita Andersson Grönlund. Ocular dimensions in relation to auxological data in a sample of Swedish children aged 4–15 years. Acta Ophthalmol. 2014 Nov; 92 (7): 682-8. doi: 10.1111/aos.12310. Epub 2014 Jan 22.
II. Lina Raffa, Eva Aring, Jovanna Dahlgren, Ann-Katrine Karlsson, Marita Andersson Grönlund. Ophthalmological findings in relation to auxological data in moderate-to-late preterm preschool children. Acta Ophthalmol. 2015 Nov;
93(7):635-41. doi: 10.1111/aos.12763.
III. Lina H. Raffa, Jovanna Dahlgren, Ann Hellström, Marita Andersson Grönlund. Ocular morphology and visual function in relation to general growth in moderate-to-late preterm school-aged children. Acta Ophthalmol. 2016 May. doi: 10.1111/aos.13085.
IV. Lina H. Raffa, Josefin Nilsson, Jovanna Dahlgren, Marita Andersson Grönlund. Electrophysiological changes in 12-year-old children born moderate-to-late preterm:
Reduced VEP amplitude and altered ERG response in MLP children. Submitted 2016.
CONTENT
ABBREVIATIONS………..…....iii
DEFINITIONS IN SHORT………...…. vi
1 INTRODUCTION……….1
1.1 Development of the Eye………...3
1.2 Systemic Impact of Moderate-to-Late Preterm Birth……….. 6
1.3 Ophthalmological Aspects of Moderate-to-Late Preterm Birth….……. 8
2 AIMS……….….. 11
3 PATIENTS AND METHODS……….…..13
4 STATISTICAL ANALYSIS ………...………. 22
5 RESULTS………. 23
6 DISCUSSION……… 43
7 CONSIDERATIONS……………….. 48
8 CONCLUSION……….. 50
9 FUTURE PERSPECTIVES……….. 55
ACKNOWLEDGEMENT……….57
REFERENCES………... 60
APPENDIX………70
ABBREVIATIONS
ACD AGA
Anterior chamber depth Appropriate for gestational age BW Birth weight
BL Birth length
BHCF Birth head circumference BMI Body mass index
CI CP
Confidence interval Cerebral palsy
CCT Central corneal thickness CT Cover test
D DA DTL EEG EOM ERG ETDRS
Diopter Dark adapted
Dawson-Trick-Litzkow Electroencephalogram Extraocular muscle Electroretinography
Early treatment diabetic retinopathy study GA Gestational age
HCF Head circumference ICD Intercanthal distance
IOP ISCEV
IUGR LA
Intraocular pressure
International Society for Clinical Electrophysiology of Vision
Intrauterine growth retardation Light-adapted
LE LT
Left eye Lens thickness
LogMAR Logarithm of the minimal angle of resolution MLP
NICU
Moderate-to-late preterm Neonatal intensive care unit NPA Near point of accommodation NPC Near point of convergence n.s. non-significant
OCT Optical coherence tomography ODA Optic disc area
pD Prism diopter
PFL Palpebral fissure length RAF
RDS
Royal air force
Respiratory distress syndrome
RE Right eye
ROP Retinopathy of prematurity
RST PPHN PRVEP
Retinal size tool
Persistent pulmonary hypertension Pattern reversal visual evoked potential SD Standard deviation
SDS Standard deviation score SE Spherical equivalent SGA Small for gestational age TAL
TTN
Total axial length
Transient tachypnoea of new-born VA
VD
Visual acuity Vitreous depth
VEP Visual evoked potential VPP Visuo-perceptual problems WHO World health organization
DEFINITIONS IN SHORT
Accommodation Ability of the eye to change the refractive power of the lens and automatically focus objects at various distances on the retina.
Amblyopia Also known as lazy eye, is decreased vision in one or both eyes due to abnormal development of vision in infancy or childhood.
Anisometropia Difference in refraction between the two eyes.
Generally a difference in power of one diopter or more of spherical equivalence is the accepted threshold for labelling the condition.
Astigmatism Refractive error as a result of unequal optical power in the two major meridians of the
cornea, which are at right angles to each other.
Cover test (CT) The main method of detecting strabismus.
Each eye is covered, while the examiner looks for movement in the non-covered eye in order to reveal any manifest strabismus.
Diopter (D) Unit of measurement of the power of a lens.
Esophoria Latent convergent strabismus, kept in check by the fusion mechanism in binocular viewing.
Esotropia Convergent strabismus (i.e. one eye deviates towards the nose).
Exophoria Latent divergent strabismus, kept in check by the fusion mechanism in binocular viewing.
Exotropia Divergent strabismus (i.e. one eye deviates away from the nose).
Emmetropia Absence of refractive error (light is focused exactly on the fovea).
Heterophoria Latent ocular misalignment, kept in check by the fusion mechanism in ocular viewing.
Heterotropia Manifest deviation not kept in check by fusion.
Hyperopia Refractive error causing the light to focus behind the retina.
front of the retina.
Ocular Motility Ability to use the extraocular muscles (EOM) to move the eye in different directions.
Premature birth Infants born before 37 weeks of gestation, late preterm are those born between 34 and 36 weeks of gestation, moderately preterm are those born between 32 and 34 weeks of gestation.
Stereo acuity Ability to perceive a three-dimensional depth, which requires adequate fusion of the images from both eyes.
Strabismus Misalignment of the visual axes.
1 INTRODUCTION
The World Health Organization (WHO) defines prematurity as birth before 37 weeks of gestation (Blencowe et al. 2012). Moderate preterm infants are defined as those born between weeks (+days) 32+0 and 33+6 of gestation, and late preterm in weeks (+days) 34+0 and 36+6 weeks of gestation (Shapiro-Mendoza & Lackritz 2012). Previously, researchers have closely studied both systemic (DʾOnofrio et al. 2013) and ophthalmological complications (Holmstrom et al. 1998) associated with extreme preterm birth (<32 weeks of gestation). Moderate-to-late preterm (MLP) infants have become the fastest-growing subgroup of preterm infants in the last decade (Verklan 2009), accounting for 84% of all preterm births (Shapiro-Mendoza
& Lackritz 2012). Preterm birth is a significant global burden, with 15.1 million babies born before 37 weeks of pregnancy every year across the world, representing one in 10 babies (Howson et al. 2013). This subset of children is becoming the focus of some researchers following the realisation of the highly increased morbidities that is associated with MLP birth, including, in the short term, respiratory morbidities, temperature and glucose dysregulation, feeding difficulties, intracranial haemorrhages, periventricular leukomalacia and infections, in addition to the long-term outcomes including neurodevelopmental, neurobehavioral sequelae and hospital readmissions (Natarajan & Shankaran 2016). Evidence is currently emerging that even near-term birth predisposes those children to a higher risk of mortality than term infants (Engle et al. 2007). To date, very few studies have focused on the ophthalmological aspects of this particular subset of children (Robaei et al. 2006; Nilsson et al. 2011).
It is well known that premature babies are at an increased risk of damage to the visual system, as well as to the cognitive and motor systems (Holmstrom et al. 1998; Holmstrom et al. 1999; O'Connor et al. 2007; Volpe 2009).
Growth retardation in utero can have subsequent negative consequences on the adult's health in the form of e.g. hormonal and metabolic effects.
Premature infants include both those whose weight is appropriate for gestational age (AGA) and those who have a low weight and/or length relative to gestational age, small for gestational age (SGA). In terms of visual function in children born preterm, several studies describe retinopathy of prematurity (ROP) with morphological retinal changes (neural and vascular), increased risk of refractive errors, strabismus, as well as visuo-perceptual problems (Hellstrom et al. 1997; Hard et al. 2000; Hellgren et al. 2016).
Recently, it has been reported that there is a relationship between birth weight (BW) and refraction at birth in both term and premature infants, with weight being a better predictor for refraction than gestational age (GA) (Varghese et
al. 2009). A study from New Zealand showed an increased risk of strabismus and subnormal visual maturation in moderate to low birth weight compared with infants of normal birth weight (Robaei et al. 2006). Studies of the growth factor IGF-I (insulin-like growth factor I) have shown that premature birth, regardless of whether an infant is born SGA or AGA, is associated with low plasma IGF-I levels in mid-childhood, suggesting partial growth hormone resistance (Cutfield et al. 2004; Mericq et al. 2005).
Researchers at the Eye clinic at the Queen Silvia Children's Hospital in Gothenburg, Sweden, have studied for many years how various factors such as growth retardation, brain damage and premature birth influence development of the eye and visual functions. Eye structures can be directly inspected and abnormalities in the eye and visual pathway can be measured using simple, non-invasive and non-painful methods. The central retinal fundus can be photographed, and objective measurement of the optic nerve, macula, nerve fibre layer and retinal vessels can be performed using a specific quantitative digital image analysis (Strömland et al. 1995; Bartling et al. 2008) and Optical Coherence Tomography (OCT). The function of the retina and visual pathways can be recorded using visual evoked potential (VEP) and full-field electroretinography (ff-ERG). The eye can thus be used as a sensitive indicator of prenatal and perinatal effects on the neural and vascular tissues.
Refraction in human infants is usually hyperopic, gradually developing toward emmetropia over the first years of life (Baldwin 1990). Most infants born hyperopic become emmetropic by age 6 to 8 years. Lewis and Maurer (Lewis & Maurer 2005) indicated that grating acuity is adult-like by 4-6 years, and letter acuity by 6 years of age. Daw (Daw 1997) stated that adult- like levels of 30 cycles per degree are reached by 3 years, and clinically, it is assumed that visual acuity (VA) is similar to an adult value of 1.0 by 5 years. This has led the authors to study the MLP children with emphasis on emmetropisation, stereoacuity and visual function by the ages of 5.5 and 8 years. Considerable maturation of the fovea, macular retinal layers (Yuodelis
& Hendrickson 1986) and blood vessels (Provis 2001) begins 24 to 27 weeks after conception and continues until early childhood (Yuodelis &
Hendrickson 1986) and as a result it was safe to consider examining the children with respect to retinal morphology and visual function by the ages of 8 and 12 years.
Animal experiments on intrauterine growth-retarded (IUGR) rats have shown that VEP activity is altered compared with normal-weight animals (Sjöström 1985). Previous studies in infants have demonstrated pathological responses to electrophysiological studies in growth-retarded infants compared with
appropriately grown infants (Thordstein et al. 2004). Maturation of VEP responses in premature infants has been shown to be more related to children's corrected GA than to their post-natal age (Roy et al. 1995).
Children with a history of ROP are reported to have reduced function of the central retina (macula) as measured by multifocal electroretinography (mf- ERG) (Fulton et al. 2005).We studied the electrophysiological changes in MLP children at the age of 12 years, when the visual system is believed to have fully matured in comparison to their full-term counterparts, as it has been noted that the rapid changes seen during infancy decrease gradually in latency in school-aged children (Brecelj 2003). However, it should also be born in mind that electrophysiological maturation might even continue into adulthood (Brecelj 2003).
This project pertains to a large proportion of children and youths growing up in Sweden today. According to the WHO, in almost all the countries with reliable data, preterm birth rates are increasing, with many survivors facing a lifetime of disability including learning, visual and hearing problems. Since brain growth and the visual system are not fully mature and thus susceptible to damage at a younger age, early prevention and treatment can have a positive effect later on in childhood. Employing methods for studying neuronal function and morphology will not only aid in providing a better understanding of the pathophysiology, but could also promote prevention, diagnosis and treatment.
1.1 Development of the Eye
Postnatal growth and emmetropisation
Refractive error is a result of discrepancy between optical refractive determinants of the eye i.e. corneal curvature, lens power and axial length.
The new-born is usually hyperopic, and within two years refractive error decreases and becomes closer to emmetropia in a process called emmetropisation. Eye growth is rapid and reaches 90% of adult proportions by approximately the age of 4. As the cornea flattens it loses refractive power, which is balanced by increasing axial length (Creig S. Hoyt 2013).
Whether this balance is guided by genetically-encoded mechanisms, environmental influences or growth factors levels, has been the subject of study for years.
Insulin-like growth factor-I during gestation is known to be crucial for the growth of most organs, but in the context of ocular affection, IGF-I is shown to influence myelination, brain development and synaptogenesis, as well as
angiogenesis (O'Kusky & Ye 2012). A study conducted on children with Silver-Russell syndrome (SRS) showed that these patients born extremely SGA and treated with growth hormone (GH) had significantly shorter total axial lengths than the controls despite a normal emmetropisation process (Grönlund et al. 2010). Parentin and Perissutti studied the effect of GH treatment on refraction and hypothesised that timely introduction could permit normal emmetropisation. The change in refraction could have been related to GH-induced somatic growth per se, or to the direct effect of GH/IGF-I (Parentin & Perissutti 2005). To our knowledge, no other group of researchers has studied the relationship between IGF-I levels and ocular growth.
Retinal anatomy and function
While funduscopic examination may reveal the fovea to appear mature soon after birth, detailed anatomic studies have shown that neither the migration of cone receptors to the foveal pit nor the movement of the ganglion cells away from the pit are complete during the first months of life. Retinal development starts in the early gestational period and by mid-gestation all retinal cells are present though very immature. It continues through the first years of life (James D. Reynolds 2011). Further development includes differentiation, migration and apoptosis of the retinal cells to form the adult retina. Fig. 1 shows the different layers of the retina. A rapid development of the electroretinography (ERG) response of both the rods and cones take place during the first four months of life and continues slowly into early school age. This process is not fully complete until several years after birth. Healthy full-term infants have a more immature ff-ERG response from rods than cones, indicating that cones mature earlier than rods. The ff-ERG response in premature infants is very immature when recorded at 30 weeks of gestation, with low amplitudes and long implicit times for both rods and cones. The ff- ERG matures continuously and, when tested in preterm infants at 40 weeks of gestation, matches the level of full-terms tested just after birth.
Electroretinography in former preterm school children has been studied only to a limited extent and mainly in children with a history of ROP (Harris et al.
2011; Akerblom et al. 2014). A normal ff-ERG response is illustrated in Fig.
2.
Figure 1. Layers of the retina.
Figure 2. Five standard electroretinography (ERG) tests with full-field stimulation.
Optic nerve myelination and visual evoked potential response
Considerable visual development occurs in the third trimester and during the first year of life. Myelination of the optic nerve and tract is incomplete at term birth and continues up to 2 years postnatally (Magoon & Robb 1981).
Although the number of cells in the primary visual cortex appears to be complete at birth, considerable increases in cell size, synaptic structure and dendritic density take place during the first six to eight years of life (Garey 1984). Visual function is dependent on VA, macular, optic nerve and primary
visual cortex functions. Visual evoked potentials are massed electrical signals generated by occipital cortical areas 17, 18 and 19 in response to visual stimulation. Pattern VEPs are elicited by abrupt contrast reversal. The transient pattern reversal VEP typically contains a small negative peak and a second negative peak as shown in Fig. 3. Visual responses have been documented in preterm infants as young as 24 weeks GA. The presence, amplitude and latency of pattern VEPs change with maturation and age. The presence of good pattern/reversal VEP response may be a good indicator of the integrity of cortical function (Aminoff 2012).
Figure 3. A Normal pattern reversal VEP.
1.2 Systemic Impact of Moderate-to-Late Preterm Birth
1.2.1 During infancy
There has been only infrequent study of MLP infants, and understanding of the developmental biology and mechanisms of disease in these infants is largely incomplete (Wang et al. 2004; Raju et al. 2006; Shapiro-Mendoza et al. 2006; Tomashek et al. 2006). Management strategies, therefore, are based on general principles, clinical experience and extrapolation from knowledge of very preterm and term infants. Even MLP infants are physiologically and metabolically immature (Kramer et al. 2000; Wang et al. 2004; Escobar et al.
2005; Oddie et al. 2005; Raju et al. 2006; Shapiro-Mendoza et al. 2006;
Tomashek et al. 2006). As a consequence, MLP infants are at a greater risk than are term infants of developing medical complications that result in higher rates of mortality and morbidity during the birth hospitalisation (Kramer et al. 2000; Shapiro-Mendoza et al. 2006; Tomashek et al. 2006).
In the last decade, improved neonatal care has increased the survival of preterm infants (Richardson et al. 1998), but unfortunately, this surge in premature births can be associated with long-term medical sequelae (Moster et al. 2008). Nevertheless, some of the reported increase in morbidity among MLP infants may be attributable to observation and detection bias, since clinicians’ thresholds for monitoring late-preterm infants for medical complications may be lower than their thresholds for term infants.
Moderate-to-late preterm birth infants have increased mortality when compared to term infants, and are at increased risk of complications including transient tachypnoea of new-born (TTN), respiratory distress syndrome (RDS), persistent pulmonary hypertension (PPHN), respiratory failure, temperature instability, jaundice, feeding difficulties and prolonged neonatal intensive care unit (NICU) stay (Gilbert et al. 2003; Ramachandrappa & Jain 2009; Verklan 2009). One study reported that new-born morbidity is seven times more likely in late preterm infants than in term infants (Shapiro- Mendoza et al. 2008). One study of all California singleton live births who survived to one year of age found that infants born at 34 to 36 weeks' gestation were 3 to 9 times more likely to require mechanical ventilation than infants born at 38 weeks' gestation (Gilbert et al. 2003; Wang et al. 2004;
Escobar et al. 2005). Moderate-to-late preterm infants are also more likely than term infants to have longer initial hospital stays and to be admitted to the NICU (Gilbert et al. 2003). Mean cost per infant was highest for children who were born 24-31 weeks ($5,393) and higher for infants born 32-36 weeks ($1,578) compared with those born at term ($725) in Massachusetts (Clements et al. 2007). Relative risk of at least one extreme event in late preterm infants is increased compared with full-term infants, and remains high until 43 postmenstrual age (Hunt 2006). Late preterm neonatal mortality rates per 1,000 live births were 1.1,1.5 and 0.5 at 34, 35, and 36 weeks respectively, compared with 0.2 at 39 weeks (McIntire & Leveno 2008).
Health care professionals have recently identified a large segment of the morbidity associated with preterm birth as being mainly attributable to MLP infants (Verklan 2009). This may be because this group is the fastest-growing sector of all preterm births, or it may simply be that they were neglected, while research was focused mainly on the complications associated with extreme premature babies. Since the MLP group is not as protected as previously believed while also taking up significant resources, it would be cost-effective to conduct formal evaluations of the therapies and follow-up strategies employed in caring for this population.
1.2.2 During childhood and adolescence
It has now been revealed that MLP infants may even have long-term neurodevelopmental consequences secondary to their late prematurity (McIntire & Leveno 2008; Moster et al. 2008). Recently, moderate preterm birth has been shown to carry considerable risks for long-term disability, lower likelihood of completing a university education, lower net income, and receipt of social security benefits (Lindstrom et al. 2007; Moster et al. 2008).
Moderate-to-late preterm birth and even marginal preterm birth (GA 37-38 weeks) also carried significantly increased risks for disability and were responsible for 74% of the total disability associated with preterm birth (Lindstrom et al. 2007). Risk of developmental delay or disability was 36%
higher among late preterm infants compared with term infants (Morse et al.
2009). Risk of suspension from kindergarten was 19% higher in late preterm infants (Morse et al. 2009). Furthermore, the presence of intelligence deficits, hyperactivity and learning disorders was reportedly more common among children with a modest low birth weight (Seidman et al. 1992; Breslau et al.
1994).
Woythaler et al (Woythaler et al. 2011) observed a higher frequency of delayed mental development at 2 years, and delayed psychomotor development in MLP compared to infants born at term. In Brazil, Santos et al. showed a higher frequency of inadequate growth at 2 years (Santos et al.
2009). Peacock et al. studied performance in regular preschool tests, comparing infants born between 32 and 37 weeks to full-term infants (Peacock et al. 2011). They found a lower frequency of good performance among preterm infants. Teune et al. found a greater risk of cerebral palsy (CP) and mental retardation (Teune et al. 2011). Moster et al. reported an increased risk of schizophrenia and a lower proportion of young individuals completing college/university (Moster et al. 2008). Teune et al. found a lower chance of completing high school (Teune et al. 2011).
1.3 Ophthalmological Aspects of Moderate-to- Late Preterm Birth
It is well known that premature babies are at high risk of developing eye complications. Until now, most studies have focused on extremely preterm infants born ≤ 32 weeks and those with extremely low birth weight (≤ 1,500 grams) because of their higher risks of mortality and serious morbidity (Ecsedy et al. 2007; Åkerblom et al. 2011; Wang et al. 2012; Åkerblom et al.
2012). Effects of extreme prematurity on ocular development are known to
include ROP, refractive error, strabismus, low visual performance, decreased contrast sensitivity, visual field defects, colour vision deficits and abnormal cognitive development (Holmstrom et al. 1998; Quinn, G. E. et al. 1998;
Holmstrom et al. 1999; Cook et al. 2008; Haugen et al. 2012). However, the impact of prematurity on visual development is not limited to the most extremely preterm infants or those with extremely low birth weights (Robaei et al. 2006). Moreover, due to the increased activity and consequent vulnerability to injury of the foetal brain during the last trimester, it is important to investigate the neurologic and visual outcomes of MLP infants carefully (Adams-Chapman 2006). To date, little is known about the effects of moderate-to-late preterm birth on the ocular and visual system.
The refractive state of the human eye is dependent on the balance of changes in overall eye size and the refractive components. Additionally, the flattening of the cornea and the decreasing power of the crystalline lens balance axial elongation in a way that maintains elongation (Zadnik et al. 2004). Any disturbance in this balance or emmetropisation mechanism may result in refractive error (Saw et al. 2004). Environmental and genetic factors, premature birth per se and the development of ROP are all known to be associated with the development of myopia (Cook et al. 2008).
A literature review revealed an increased prevalence of behavioural and emotional problems in very low birth weight or preterm infants, with rates ranging from 25–55%, as opposed to 7% in controls (Hayes & Sharif 2009). Children born very preterm (˂32 weeks) and without major neurodevelopmental sequelae have an increased prevalence of ophthalmic impairments at primary school age that are associated with visuo-perceptual, motor and cognitive defects (Hard et al. 2000; Cooke et al. 2004).
Researchers at the eye clinic at the Queen Silvia Children’s Hospital in Gothenburg, Sweden, have a long history of studying the influence of various factors such as growth retardation, brain damage and preterm birth on the development of the eye and visual function (Grönlund et al. 2004; Grönlund et al. 2006; Martin et al. 2008). It has previously been found that a persistent reduction in serum levels of insulin-like growth factor-I (IGF-I) after birth is associated with subsequent poor angiogenic development (Hellström et al.
2003). Retinopathy of prematurity development was lower in infants that were placed in early aggressive parental nutrition inducing high IGF-I, as an inverse correlation is found between ROP and IGF-I levels (Can et al. 2013).
At our department, it has previously been found that VEP latencies did not differ in SGA children compared with AGA children in the MLP children studied (Nilsson et al. 2011), supporting previous studies (Scherjon et al.
1996) suggesting a catch-up in neurophysiological maturation during the first year of life compared with controls.
The increased survival of prematurely born infants poses a long-term problem in terms of increased incidence of ophthalmological problems such as strabismus, amblyopia and refractive errors (Schalij-Delfos et al.
2000). These findings have potentially important implications for the follow- up of MLP preterm-born children, and therefore require confirmation in large population-based studies encompassing these preterm children.
2 AIMS
Overall aim
• Identifying visual function, eye morphology and growth in children born MLP (SGA and AGA), comparing the findings to their full-term counterparts and relating the data to other growth parameters and IGF-I levels.
• Identifying "normal" variations of the eye and vision (function and morphology) in children aged 4-15 years in relation to auxological data.
Paper I
The purpose was to characterise normal growth patterns of ocular components and to relate them to auxological data in a sample of Swedish children aged 4–15 years.
Paper II
To evaluate ophthalmological findings in preschool children born MLP (SGA and AGA) and relate the findings to auxological data and IGF-I levels at birth and at 5.5 years of age.
Paper III
To study ocular morphology and visual function in relation to other growth parameters in children born MLP at 8 years of age.
Paper IV
To study electrophysiological changes in relation to fundus morphology in children born MLP at 12 years of age.
Research Questions
1. Is there a difference in the eye and visual function between children born MLP (SGA and AGA) and healthy term infants?
2. Do children born MLP have normal eye growth and hence normal refraction development (emmetropisation) at the age of eight years?
How is their eye growth and visual maturation in relation to the other growth parameters?
3. Is there a difference in retinal function and morphology between children born MLP and healthy full-term children?
4. What is the relationship between circulating IGF-I levels and ocular and visual maturation?
5. What is the "normal" variation in ophthalmologic variables in healthy term infants at 4-15 years of age in relation to other growth parameters?
3 PATIENTS AND METHODS
3.1 Patients Paper I
At the Department of Paediatric Ophthalmology at the Queen Silvia Children’s Hospital, Gothenburg, Sweden, a total of 143 children (67 girls;
76 boys) were evaluated with regard to ophthalmological and auxological variables. The children were selected from four different schools and three day-care centres in the Gothenburg area to represent a city centre, a suburban, and a rural area in order to reflect the socioeconomic mix of the area from which the cohort study was drawn. Inclusion criteria were age between 4 and 15 years, birth in Sweden, and residence in the Västra Götaland region, Sweden. Any history of ocular or serious health diseases were regarded as exclusion criteria. The aim was to recruit at least five girls and five boys from each year of age. A detailed description of the population, including medical, ethnic and socio-economic background, as well as ophthalmologic findings, (Grönlund et al. 2006), in addition to the orthoptic evaluation (Aring et al.
2005) and fixation (Aring et al. 2007), has previously been presented.
Papers II-IV
In an ongoing, prospective, population-based study, all children born MLP (GA 32 weeks+0 days to 36 weeks+ 6 days) between 2002 and 2004 in either of the two available maternity wards in Gothenburg, Sweden (Östra and Mölndal University Hospitals) were invited, through their parents or guardians, to participate (Fig. 4). Children with syndromes, chromosome abnormalities, severe malformations or mothers with severe chronic diseases were not included. None of the participating children had any previous history of ROP. Guardians of 247 potentially eligible children (110 girls and 137 boys) agreed to allow them to join the neonatal study. At 5.5 years of age, all children still included in the study (n=127) were invited to participate in an ophthalmological investigation. Of these 127 MLP children, 78 MLP children with a GA of 32–36 weeks (34 girls; mean age 5.7 years; 21 SGA and 57 AGA children) agreed, with their guardians’ permission, to take part in this sub-study focussing on VEP studies (Nilsson et al. 2011) and detailed eye examinations including orthoptic evaluation, which have been related to the different growth parameters obtained at birth and at 5.5 years of age
(Paper II). The data was compared with age- and sex-matched controls born full term from our reference group in Paper I (n=35).
At 8 years of age, all children included in the ophthalmological part of the study (n=78) at 5.5 years (Paper II) (Raffa et al. 2015) were invited to participate in this sub-study focusing on eye growth, refraction development and retinal morphology in relation to other growth parameters. Of these 78 MLP children, 50 (29 boys, 21 girls) with a mean ± standard deviation (SD) GA of 35±1.5 weeks and mean birth weight (BW) of 2299±469 g agreed, with their guardians’ permission, to take part in the study (Paper III).
Examinations were carried out on 12 children born SGA (7 boys, 5 girls;
mean±SD GA = 34±1.4 weeks; BW 1843±420 g) and on 38 children born AGA (22 boys, 16 girls; mean GA±SD = 35±1.5 weeks, and BW = 2443±388 g). Forty-three children were recruited by invitation through local schools in different areas of Gothenburg to serve as controls. Inclusion criteria for the control group included children born at term aged 8 from these randomly chosen local schools with no previous history of ocular or systemic diseases.
All 50 examined in the previous study at 8 years of age (Paper III) (Raffa et al. 2016) were invited, through their parents or guardians, to participate in the last sub-study (Paper IV). Examinations were carried out on 22 children (11 boys, 11 girls; mean±SD GA 34.5±1.7 weeks; BW 2266±482 g) who agreed to participate. Twenty-one children were recruited from local Gothenburg schools to serve as controls. Inclusion criteria for the control group included children born at term (GA≥37 weeks) with no previous history of ocular or systemic diseases.
Figure 4.Flowchart showing a summary of children’s participation in this study.
AGA = appropriate for gestational age; n = number of individuals; SGA = small for gestational age.
Neonatal study n=247 (2002-2004) 5.5 year Follow-up n=127 PAPER II (2002-2004) Eye Examination n=78 SGA=21 AGA=57 8 year follow-up n=78 PAPER III (2002-2004) Eye Examination n=50 SGA=12 AGA=38 12 year follow-up n=22 PAPER IV (2002-2004)
Refusals n= 2 Failed to respond n=26 Full-term controls n=21 Refusals n=16 Failed to respond n= 12
Full-term controls n=43 Refusals n=30 Failed to respond n=16 Copies of medical records only n=3
Sex- and age- matched controls n=35 Reference group n=143 PAPER I
Neonatal subset n=167 Exmined n=78 (2002-2003) 10 year follow- up n=78 (2002-2003) Eye Examination n=33 SGA=9 AGA=24
Refusals n=3 Failed to respond n=37 Moved n=5 Sex- and age- matched controls n=28
Allvin K et al 2014Nilsson J et al 2011
3.2 Methods
Paediatric Examinations 3.2.1 Auxological data
Weight, length and head circumference (HCF) at birth and at time of assessment were measured in accordance with a standard protocol and converted into standard deviation scores (SDSs) based on Swedish reference values (Niklasson et al. 1991). Body mass index (BMI) was computed as weight (kg) divided by height squared (m²).
3.2.2 IGF-I data
IGF-I was analysed using the new automated immunoassay (IDS-iSYS;
Immunodiagnostics Systems) and transformed to SDS at birth (Bidlingmaier et al. 2014), 5.5, 8 and 10 years of age (Lofqvist et al. 2001). Calculating the delta IGF-I, values from cord serum IGF-I at birth were subtracted from the serum values at time of investigation.
Ophthalmic Examinations
3.2.3 Visual Acuity tests for far and near fixation
Visual acuity was tested using a linear KM-Boks chart. If a child could not manage to read the KM-Boks chart, an HOTV chart was used (Moutakis et al. 2004). Distance vision was tested monocularly at a distance of 3 m, and near vision was tested binocularly at a distance of 0.33 m. Values were noted in Snellen decimal format and converted to logarithm of minimal angle of resolution (logMAR) (Appendix 1). Amblyopia was defined as a difference in VA between the eyes in at least two lines with VA in the amblyopic eye
≤0.2 logMAR that could not be explained by structural abnormalities in the eye. Subnormal VA was defined as VA<0.2 logMAR in either eye.
3.2.4 Refraction under cycloplegia
This was performed using an autorefractor (Topcon A6300; Topcon Corporation, Tokyo, Japan) following a single instillation of a mixture of cyclopentolate (0.85%) and phenylephrine (1.5%). Significant refractive errors were defined as a spherical equivalent (SE) of myopia ≥0.5 dioptres
(D) or hyperopia ≥2.0 D. Astigmatism was assessed at a level of ≥1.0 D and anisometropia at ≥1.0 D SE.
In addition to the examinations mentioned above, the following tests were performed on the children:
Paper I
3.2.5 Assessment of ocular dimensions
The inner canthal distance (ICD) and right and left palpebral fissure lengths (PFL) were measured in mm using a ruler (Hall et al. 1989). Canthal index was calculated as ICD x 100/outer canthal distance as a percentage (Taylor &
Hoyt 2005). Total ocular axial length (TAL), anterior chamber depth (ACD), lens thickness (LT), and vitreous depth (VD) were measured by means of ultrasound biometry (Paxis Version 2.01; BIOVISION, Clermont Ferrand, France). The mean value of ten measurements of each eye was recorded where possible.
3.2.6 Photography of the ocular fundus for qualitative image analysis
The children underwent cycloplegia and ocular fundus photographs were obtained using a Topcon 50-VT fundus camera (Topcon Corporation, Tokyo, Japan). Only well-focused pictures were accepted. The optic disc was identified as the inner margins of the nerve tissue excluding the white scleral rings. The cups were identified manually by contour and pallor in both eyes.
The optic disc area (ODA) and cup areas were measured by marking their outlines with a cursor. The projected areas were analysed using a specifically designed computer-assisted digital mapping system (Strömland et al. 1995).
Cup-disc ratios and the neuro-retinal rim were calculated.
Paper II
3.2.7 Orthoptic evaluation Strabismus and motility
Heterotropia was diagnosed using a cover test with fixation at a distance of 3 m and 0.33 m respectively, and was defined as a permanently or intermittently manifested deviation. The angle was measured in prism dioptres (pD) using alternate and prism cover tests. The cover test (CT) was
performed at 3.0 and 0.3 m. During the Hirschberg test and the CT, any anomalous head postures and presence of nystagmus were noted. The nomenclature was eso-, exo-, hypo- and hypertropia (Von Noorden &
Campos 2002). Motility (versions and ductions) was tested using a penlight in the nine positions of gaze. Significant misalignment was defined as heterotropia at any distance, or exophoria (-) as values below the 5th percentile in the control group, and esophoria (+) as values above the 95th percentile. Cut-off values at distance -5 prism diopters (PD) to +3PD and at near -11PD to +7PD (Aring et al. 2005).
Stereo acuity testing
The TNO test was used to evaluate stereo acuity (Von Noorden & Campos 2002). For the TNO test, stereo acuity was defined as the smallest level of disparity at which both test figures were correctly identified, and it was considered reduced if it was more than 60 seconds of arc (Aring et al. 2005).
Near point of convergence
The Royal Air Force (RAF) ruler was used to measure the near point of convergence (NPC). A mean value of three measurements was recorded. The near limit is 6 cm (Von Noorden & Campos 2002).
Near point of accommodation
The near point of accommodation (NPA) was measured binocularly in D as the target was brought toward the patient with the RAF ruler, using the push-up test (Scheiman & Wick 2008).The cut-off limit was set at 20 D according to Duane’s standard curve of accommodation (Duane 1922). A mean value of three measurements was recorded.
3.2.8 Examination of the anterior segment, media, and ocular fundus
Examination of the anterior segment of the eye was performed using a slit lamp, and the ocular fundus was examined by means of indirect ophthalmoscopy. Fundus photographs were taken and inspected by the observers for optic nerve, macular or vessel pathologies.
3.2.9 History of visual perception
A structured history-taking was conducted with respect to visuo-perceptual problems (VPPs) in five different areas, these being recognition, orientation, perception of depth and motion, and simultaneous perception (Dutton et al.
1996). The history-taking was based on 12 selected questions regarding visual perceptual cognitive problems translated into Swedish. Whereas questions 1-4 dealt with recognition, questions 5-7 addressed orientation, question 8 asked about perception of depth, questions 9 and 10 about perception of motion and questions 11 and 12 about simultaneous perception (Appendix 2).
Paper III
3.2.10 Ocular dimensions
The ICD and the right and left PFLs were measured in mm using a ruler (Hall et al. 1989). Canthal Index was calculated as ICDx100/outer canthal distance as a percentage. Total ocular axial length was measured by means of ultrasound biometry (IOL master, 500 Zeiss Meditec, Jena, Germany). The mean value of 10 measurements (when possible) of each eye was recorded.
3.2.11 Intraocular pressure
Intraocular pressure (IOP) was measured using a handheld tonometer (TA0li, I care, Finland Oy, Espoo, Finland) which is automated to take the average of six readings. Readings with high deviations were discarded and repeated.
3.2.12 Central corneal thickness
Central corneal thickness (CCT) was measured twice – firstly using a handheld ultrasound pachymeter (Pocket Class II; Quantel Medical Inc, Clermont-Ferrand, France) with the mean of five measurements being taken in each eye, and secondly using the Spectral Domain (SD)-OCT machine.
3.2.13 Fundus photography
Ocular fundus photographs were obtained using the fundus camera (Nonmyd- 7 Kowa, Tokyo, Japan) and were taken in cycloplegia. Digital fundus photographs were fed into the RetinaSizeTool (RST) program (Bartling et al.
2008). Optic disc area, neuroretinal rim area and cup area were analysed by
manually marking the outlines of the optic disc and cup in addition to the fovea, yielding the projected area results (Bartling et al. 2008).
Paper III-IV
3.2.14 Optical coherence tomography variables
Retinal and papillary nerve fibre layer thickness and retinal thickness were measured by analysing images taken with the SD-OCT, using 3D-OCT 1000 software from Topcon, Tokyo, Japan, to obtain the average thickness of 9 early treatment diabetic retinopathy study (ETDRS) sectors (i.e. sectors A1 to A9: central (C), inner superior (IS), outer superior (OS), inner inferior (II), outer inferior (OI), inner nasal (II), outer nasal (ON), inner temporal (IT), and outer temporal (OT)), and the retinal nerve fibre layer (RNFL) thickness, foveal minimum, and total macular volume in the sectors A1 to A9 (Fig. 5).
The children were asked to fixate on the internal target and the scans automatically provided the aforementioned results. Imaging of dilated fundi was obtained in a dim room, with the highest quality image selected for analysis. The software also calculated average thickness for each RNFL macular segment in the 9 ETDRS sectors. The papillary RNFL thickness was measured in four sectors (superior (S), inferior (I), nasal (N) and temporal (T)) and an average thickness was automatically calculated.
Figure 5. Example of macular thickness measurements obtained using the Topcon SD-Optical coherence tomography(OCT) system. OCT image of a healthy subject (A) Fundus photograph of a healthy subject; the box indicates a 6×6 mm scanning area using 3D macular protocol (B). Image of standard early treatment diabetic retinopathy study (ETDRS) map (C), showing map diameters centred on fovea (left) and 9 standard ETDRS regions (right).
